T. Kvalseth. The American Statistician, 39 (4):
pp. 279-285(1985)
Abstract
The coefficient of determination (R2) is perhaps the single most extensively used measure of goodness of fit for regression models. It is also widely misused. The primary source of the problem is that except for linear models with an intercept term, the several alternative R2 statistics are not generally equivalent. This article discusses various considerations and potential pitfalls in using the R2's. Specific points are exemplified by means of empirical data. A new resistant statistic is also introduced.
Description
JSTOR: The American Statistician, Vol. 39, No. 4 (Nov., 1985), pp. 279-285
%0 Journal Article
%1 1985
%A Kvalseth, Tarald O.
%D 1985
%I American Statistical Association
%J The American Statistician
%K model r2 regression statistics
%N 4
%P pp. 279-285
%T Cautionary Note about R2
%U http://www.jstor.org/stable/2683704
%V 39
%X The coefficient of determination (R2) is perhaps the single most extensively used measure of goodness of fit for regression models. It is also widely misused. The primary source of the problem is that except for linear models with an intercept term, the several alternative R2 statistics are not generally equivalent. This article discusses various considerations and potential pitfalls in using the R2's. Specific points are exemplified by means of empirical data. A new resistant statistic is also introduced.